THE CONSTITUTION, the LABOR UNION, and "GOVERNMENTAL ACTION" HARRY H.Wellingtont

THE YALE LAW JOURNAL

VOLUME 70 JANUARY 1961 NUMBER 3

THE CONSTITUTION, THE LABOR UNION, AND "GOVERNMENTAL ACTION" HARRY H.WELLINGTONt

LEE Oliphant could not join the Brotherhood of Locomotive Firemen and Enginemen. Yet under the Railway Labor Act 1 this union represented him in negotiations with his employer. The Brotherhood took to membership white locomotive-firemen; by constitutional provision it excluded all others.2 Oliphant was by occupation a locomotive-fireman on the railroad; by race, he was a Negro. Undeterred by the shibboleth that the law cannot compel the spirit of brotherhood, Oliphant asked a federal district court to order his admission into the union. He argued that the due process clause of the fifth amendment to the United States Constitution requires no less. This should cause raised eyebrows for one important reason. The fifth 3 amendment is a limitation only upon the actions of the federal government. Certainly the actions of a labor union are not ordinarily considered those of the government in Washington. 4 Ostensibly private actions, however, may occasionally have a sufficient nexus with governmental action to justify use of the Constitution as an instrument of control. If Oliphant could show such a nexus, it is absolutely clear that his suit would succeed. 5 Oliphant failed to make such a showing.6 Yet the facts of his case may be susceptible to an anal- tProfessor of Law, Yale University. 1. 44 Stat. 577 (1926), as amended, 45 U.S.C. §§ 151-88 (1958). 2. The Brotherhood's "Constitution limits membership to applicants 'white born'". Oliphant v. Brotherhood of Locomotive Firemen and Enginemen, 262 F.2d 359 (6th Cir. 1958), cert. denied, 359 U.S. 935 (1959). 3. "[T]he First and Fifth Amendments to the Constitution of the United States .... apply to and restrict only the Federal Government and not private persons. See Corrigan v. Buckley, 271 U.S. 323, 330; Talton v. Mayes, 163 U.S. 376, 382, 384; Withers v. Buck- Icy, 20 How. 84, 89-91; Barron v. The Mayor and City Council of Baltimore, 7 Pet. 243; see also Virginia v. Rives, 100 U.S. 313, 318." Public Util. Comm'n v. Pollak, 343 U.S. 451, 461-62 (1952). 4. 'We do not suggest that labor unions which utilize the facilities of the National Labor Relations Board become Government agencies or may be regulated as such." Ameri- can Communications Ass'n, CIO v. Douds, 339 U.S. 382, 402 (1950). 5. Cf. Brown v. Board of Educ., 347 U.S. 483 (1954) ; Gayle v. Browder, 352 U.S. 903 (1956) ; Holmes v. City of Atlanta, 350 U.S. 879 (1955) ; Muir v. Louisville Park Theatrical Ass'n, 347 U.S. 971 (1954). 6. Oliphant v. Brotherhood of Locomotive Firemen & Enginemen, 156 F. Supp. 89 (N.D. Ohio), cert. denied, 355 U.S. 893 (1957), aff'd, 262 F.2d 359 (6th Cir. 1958), cert, THE YALE LAW JOURNAL [Vol. 70: 345

ysis sufficient to satisfy the governmental action requirements of the fifth amendment. The Supreme Court of the United States will one day surely be called upon to decide this question. While Lee Oliphant's case is recent, it is by no means a unique addition to the law reports.7 In learned journals, monographs, and books, issues analogous to those raised by Oliphant for some time now have been extensively noted.8 Some of this literature goes far beyond the question of whether the fifth amendment may be used to compel admission of a Negro employee to a railroad brotherhood. Ranging wide through society and deep into the Constitution commentators have suggested that all or most "powerful" private groups should be subject to all or most provisions of the Constitution. 9 The business corporation and the labor union have been the principal targets of these suggestions, and the Bill of Rights and the fourteenth amendment have been envisioned as the principal instruments for control.10 For example, we are told that: The corporate organizations of business and labor have long ceased to be private phenomena. That they have a direct and decisive impact on the social, economic, and political life of the nation is no longer a matter of argument. It is an undeniable fact of daily experience. The challenge to the contemporary lawyer is to translate the social transformation of these organizations from private associations to public organisms into legal terms. 1 denied, 359 U.S. 935 (1959) ("In view of the abstract context in which the questions sought to be raised are presented by this record, the petition for writ of certiorari . . . is denied."). 7. Compare, e.g., Ross v. Ebert, 275 Wis. 523, 82 N.W.2d 316 (1957), with Betts v. Easley, 161 Kan. 459, 169 P.2d 831 (1946). 8. See, e.g., HALE, FREEDOM THROUGH LAW (1952); Miller, Private Governments and the Constitution (Occasional Paper for Center for the Study of Democratic Institutions, 1959) ; Rauh, Civil Rights and Liberties and Labor Unions, 8 LA. L.J. 874 (1957) ; Sum- mers, The Right to Join a Union, 47 COLUM. L. REv. 33 (1947). 9. See, e.g., Miller, The Constitutional Law of the Security State, 10 STAN. L. REv. 620, 655-56 (1958) ; Ming, Racial Restrictions and the Fourteenth Amendment: The Re- strictive Covenant Cases, 16 U. CHI. L. Rav. 203, 235-38 (1949) ; Malick, Toward a New Constitutional Status for Labor Unions: A Proposal, 21 RocKy MT. L. REv. 260 (1949) ; PEKELiS, LAw AND SOCIAL AcTION 91-128 (1950); Latham, The Commonwealth of the Corporation,55 Nw. U.L. Rv. 25 (1960). "[W]e could ... consider applying to the corporation the whole pattern of controls laid upon the states when the Federal Republic was created under the Constitution of 1787. The parallel is not too fanciful since many of the states in fact started out as corporations created by the Crown, bodies politic endowed with public authority." Id. at 35. (Emphasis in original.) 10. See, e.g., Berle, Constitutional Limitations on Corporate Activity-Protection of PersonalRights from Invasion Through Economic Power, 100 U. PA. L. REv. 933 (1952) ; BERLE, ECONOMIC POWER AND THE FREE SocIETY 17-18 (Fund for the Republic, 1957); Rauh, Civil Rights and Liberties and Labor Unions, 8 LAB. L.J. 874 (1957). 11. Friedmann, Corporate Power, Government by Private Groups, and the Law, 57 C9=4. L, REv, 155, 176 (1957). 1961] GOVERNMENTAL ACTION

Another commentator suggests that: As a beginning, we can set out the following propositions: (1) The Con- stitution was framed on the theory that limitations should exist on the formal exercise of power in government but not on control exercised un- officially. (2) The essential problem of individual liberty, however, is one of freedom from arbitrary restraints and restrictions, wherever and however imposed. (3) The Constitution should be so construed as to apply to arbitrary applications of power against individuals by centers of private government. (4) The main flow of group decisions in the factory community would not be thrown into litigation or controversy by such a constitutional construction, but only those which directly and substantially affect an individual. (5) It would take only a slight modification of present constitutional doctrine to effect such a cofistitutional construction.' 2

These are engaging ideas which at the simplest level pose two questions: (1) Why the emphasis on the business corporation and the labor union? (2) Why the choice of constitutional provisions as the means of regulating these institutions ? The answer to both questions starts with a commonplace: some business corporations and some labor unions are big-indeed enormous, in every sense of the word. Bigness suggests power. And big business and big labor have power. [B]usiness and labor currently exercise vast powers. First, they have power over the millions of men and women whose lives they largely control as employees or as members. Second, they exercise power more in- directly, though not less powerfully, over the unorganized citizens whose lives they largely control through standardized terms of contract, through price policy, through the tempo of production and the terms and conditions of labor. Last, they exercise control over the organized community, represented by the organs of the state, in a multitude of ways: direct lobby pressures, control over the election and policies of the elected representatives of the people, control over the appointment of the judiciary in many states, and far reaching control over the mass media of communication.' 3 This recital of the way in which unions and corporations may use various types of power 14 suggests certain similarities between big unions, big corporations, and big government. It suggests how these private institutions touch upon our social, economic, and political life. Only government, and certainly

12. Miller, supra note 8, at 12. 13. Friedmann, supra note 11, at 176-77. 14. Professor Manning on "Power" needs to be quoted again and again. The Corporate Power thesis conceives of Power as though it were Mercantilist gold bullion-physically piled on someone's desk, infinitely fungible unit for unit, and indifferently expendable to achieve any result. Yet all our experience is squarely to the contrary. Power to do A is not power to do B. The management of a particular company may be so free of shareholder control that it can pay itself salaries beyond the dreams of avarice. Here is power indeed; but what may be inferred from it as to the management's power to do other things? Can it control prices?-elect a Senator-prevent its workers from voting-secure the passage of a constitutional THE YALE LAW JOURNAL [ Vol. 70 :345 no other private group, has a comparable impact on the shape of society; and, of course, this is the reason labor unions and business corporations have been singled out for special attention in the literature. 15 It is also one reason why the Constitution comes to mind as an instrument for control. Unions and corporations have an importance in our lives which the founding fathers would have thought possible only of government itself.' 6 Another reason some turn to the Constitution as the instrument for control- ling these institutions is the formal structure of unions and corporations. At the formal level these institutions look much like government.' 7 Such emotional phrases as "industrial democracy," "union democracy," and "shareholder democracy" seem to reinforce the comparison and to bolster the "looks-like- government" school of thought. Undue fascination with the supposed structural and power similarities unions and corporations have with government can be misleading. Some commentators have become so fascinated, and it is this fascination that provides the decisive reason for their advocacy of regulation by the Constitution. There is also little doubt that some commentators are motivated by an undisciplined desire to "let the mind be bold." This sort of thinking deserves little sym- pathy. The Bill of Rights and the fourteenth amendment are the great instruments with which courts protect the people from misused governmental power. The view that because unions and corporations are somehow similar to government they too should be restrained by these same constitutional provisions has perhaps an aesthetic and emotional appeal. Its analytical shortcom- ings, however, are fatal. The need to regulate unions and corporations is undeniable; but it need not be assumed a priori that the Constitution is the proper regulatory instrument. Other, more appropriate, means may be available to accomplish the same desirable ends.

25% income tax limit-beat a union strike-or even stop dividend payments? We don't know. On the other hand, a management that can muster the economic leverage to peg a market price may or may not be under iron stockholder control, may or may not have any substantial political influence, may or may not be at the mercy of a stronger union, and may or may not have to turn over the bulk of its earnings to a retired patentee rocking on his veranda at Cannes. We cannot deduce any of these facts. We can only ask the questions one at a time, then go find out. Except as poetry, 'Power' is not usable in the gold bullion sense. Power becomes a usable analytic conception only when used in a complete statement-a statement of someone's power to do something, affecting someone in some way. Without an infinitive, 'Power' is a bell without a clapper. Manning, CorporatePower and Individual Freedom: Some Geieral Analysis and Partic- ular Reservations, 55 Nw. U.L. Rav. 38, 44-45 (1960). 15. "When any villain seems to threaten freedom in our economy the most popular suspects are big business, big labor, and big government. The known presence of a giant is a source of apprehension not only to the pygmies but also to other giants whose power may be challenged." Hoffman, Freedom in The Modern American Economy: A Sym- posium Introduction, 55 Nw. U.L. REv. 1 (1960). 16. See text at notes 11-12 supra. 17. See DANKERT, CONTEMPORARY UNIONISM: (1948); Latham, The Body Politic of the Corporation,in MAsoN, THE CORPORATIO INrMODERN SocimrY 218 (1959). 19611 GO VERNMENTAL ACTION

If a union or a corporation is in interstate commerce Congress can pass regulating legislation.1 8 State legislatures also have the power to control, unless federal law has occupied the field. 19 Of course, it is true that there may be a lag between an observed need to regulate particular corporate or union conduct and the ability of legislating institutions to accomplish the needed reform. For the reformer the most apparent sources of regulation are then the Constitution and the courts. This thinking is attractive; but it invites support only if on a balance of relevant factors it appears that the gain is not overweighed by the violence done to our traditional institutions and the roles they perform in society. In making this evaluation it should be noted that the power of business corporations and labor unions to do a great many things will not be affected by 20 the imposition of constitutional restraints. Constitutional restraints could be effective limitations on union or corporate power in areas involving the treat- 2 1 ment of employees and of those seeking work in an enterprise. With respect

18. See, e.g., NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1 (1937). 19. See, e.g., San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236 (1959). 20. But see Latham, The Commonwealth of the Corporation,55 Nw. U.L. REv. 25, 35-37 (1960): If [the Constitution] ... were to be applied to the corporations, the Federal Gov- ernment should guarantee to every one of them a 'republican form of government' .... Interstate compacts may be made with the consent of the Congress, and inter- corporation compacts in a United Corporate States of America would be subject to the same rule .... Some of the other provisions of the Constitution, if applied to the corporations, might have startling effects and some would have none. What of the privileges and immunities clauses or the full faith and credit clause? The privileges and immunities of citizens of the United States that no corporation would be permitted to abridge would possibly create no new forms of action, but article IV, section 2 says that the citizens of each state shall be entitled to all the privileges and immunities of citizens in the several states. Does this mean that stockholders of General Motors should be given free access to the public facilities of duPont or General Electric? Can executive personnel of AT&T demand keys to the more private precincts usually reserved for the executive personnel of Metropolitan Life? And what of section 4 of article IV, promising to protect each of the states against invasion? Could Sewell Avery in a switch of attitudes appeal to the Federal Gov- ernment for aid in beating off the invasion of Montgomery Ward by Louis Wolfson? Other provisions of the Constitution might deprive the corporations of certain privileges they now enjoy. Section 10 of article I would forbid them to remit bills of credit and so centralize-nationalize-all banking in the country. Nor could they grant titles of nobility, a blow against fraternal orders, surely. And they would all be disarmed, having lost to the Federal Government the power to raise and maintain armies. This would presumably put Pinkerton's and Brink's out of business. But the taxing power of the corporation, its power to take sums from the buyer by manip- ulating administered prices, would be concurrent, and its internal police power would remain intact, although limited to enactments in the interests of the safety, health, morals, and welfare of those under its jurisdiction and authority. 21. Most of the problems which come to mind concern fair representation and internal union affairs. And this means that here application of the Constitution is often likely to be of only academic importance. Federal statutes regulate both of these areas up to or beyond THE YALE LAW JOURNAL [Vol. 70: 345 to their general economic activities and to their relations with government- except to the extent that these problems overlap employee treatment--control over unions or corporations by constitutional prohibitions will be relatively in- effective.2 2 Because of the role performed by the labor union in an enterprise- it is the representative of the bulk of employees-the focus here will be on unions. This focus is especially propitious since, with few exceptions, the litigation in this area of law has involved labor unions.2 3 Examination of the present state of the law is an appropriate place to begin.

DOCTRINE Lee Oliphant's problem is a useful vehicle for this examination. To compel his admission into the union by force of the Constitution, Oliphant would have to demonstrate that there is some doctrinal nexus between the fifth amendment's due process clause or the fourteenth amendment's equal protection clause and his exclusion from the Brotherhood of Locomotive Firemen and Enginemen because of his race. The trick is to implicate the government in some way and make it a partner to the union's deed. Governmental action must be established if Oliphant's exclusion is to be tested-and surely found unreasonable-by the requirements of the Constitution.2 4 For the moment we put to one side the possibility that the mass of federal 25 legislation touching the labor union can supply this needed nexus.

I If Oliphant had brought his suit in a state court of general jurisdiction, might not the state government be sufficiently implicated merely by the court's denial of his prayer? To deny relief, the argument might run, is tantamount to the state government sanctioning the Brotherhood's exclusion of Negroes. The due process and equal protection clauses of the fourteenth amendment, which apply to state action, would be brought into issue. In the state tribunal Oliphant's position would be either that the Brotherhood's behavior violates state law, or that if it is sanctioned by state law a partnership between state and brotherhood is established thereby, and accordingly exclusion must be judged by fourteenth amendment standards. At common law state courts will not order unions to admit applicants to membership. 26 This "hands-off" policy is state policy. States, of course, can a constitutional requirement. See text at notes 64-74, 95-103 infra. The principal non- academic problems are the ones discussed in this paper. 22. See Manning, supra note 14, at 38. But see Berle, supra note 10, at 933. 23. The principal exception is Marsh v. Alabama, 326 U.S. 501 (1946). See text at note 62 infra. 24. See note 5 supra. 25. See text at notes 54-62 infra. 26. E.g., Miller v. Ruehl, 2 N.Y.S.2d 394, 166 Misc. 479 (Sup. Ct. 1938). Thorman v. International Alliance of Theatrical Stage Employees, 49 Cal. 2d 629, 320 P.2d 494 (1958), vacating, 307 P.2d 1026 (Cal. Dist. Ct. App. 1957), is the single exception. In 1961] GOVERNMENTAL ACTION

say that exclusion because of race is illegal, and some fair employment practice acts do so provide.2 7 While it might not be strictly appropriate to characterize the traditional "hands-off" policy as a sanctioning of union discrimination, it surely is proper in such cases to say that state law tolerates the discrimination. But a state policy of toleration is not, without more, state action within the meaning of the fourteenth amendment. If it were, the amendment would apply to the behavior of social fraternities, tennis clubs, and clubs of all kinds.2 8 In- deed, most behavior would in the end be state action for fourteenth amendment purposes. That no serious suggestion of this kind has ever been raised is perhaps sufficient comment on its undesirability. Since the Civil Rights Cases of 1883 29 the Supreme Court has consistently interpreted the amendment to restrict its scope to an area of affirmative state action and a slight penumbrum beyond. Shelley v. Kraemer 30 is the leading decision. Shelley, a Negro, was the vendee of land encumbered by a covenant providing that "no part of said property or any portion thereof shall be ...occupied by any person not of the Caucasian race. . . -31 The owners of other property subject to this restric- tion sued in a Missouri court to enforce the agreement. They sought to restrain Shelley from taking possession of the property and to divest him of title. The Supreme Court of Missouri enforced the agreement, and granted the relief sought. On appeal the Supreme Court of the United States held that in so do- ing the Missouri court had transgressed the equal protection clause. Shelley built on two propositions that were clear from prior authority. First, a state, whether by statute or common law, cannot itself restrict the use of property on racial grounds.32 Second, the acts of a state court are the acts of the state so far as the fourteenth amendment is concerned. 33 Accordingly, the only new issue-decided affirmatively in Shelley v. Kraemer-was whether enforcement of the private covenant, which restricted the use of property on the basis of race, was sufficient state action to bring the equal protection clause into play. Can it be said that the action of the state in Shelley v. Kraerner is action in any sense different from that hypothesized in Lee Oliphant's case, in which the state tolerated the Brotherhood's discrimination through the refusal of its court to order Oliphant's admission? This issue was resolved by the Court in Shelley: many jurisdictions the closed union, closed shop, will not be tolerated. The union is given the choice of admitting the plaintiff-employee to membership or not enforcing its union security contract to the plaintiff's disadvantage. For the California law on this, see Wil- liams v. International Bhd. of Boilermakers, 27 Cal. 2d 586, 165 P.2d 903 (1946) ; James v. Marinship Corp., 25 Cal. 2d 721, 155 P.2d 329 (1944) ; cf. Riviello v. Journeyman Bar- bers Union, 88 Cal. App. 2d 499, 199 P.2d 400 (Dist. Ct. App. 1948). 27. E.g., N.Y. EXECUTinv LAW § 296(1) (b) ; N.Y. Civ. RIGHTs LAW § 43. 28. In other words, an applicant for membership, excluded for an arbitrary reason from a club, could gain admission by force of the fourteenth amendment. 29. 109 U.S. 3 (1883). 30. 334 U.S. 1 (1948). 31. Id. at 4-5. 32. Buchanan v. Warley, 245 U.S. 60 (1917). 33. See Shelley v. Kraemer, 334 U.S. 1, 14-18 (1948). THE YALE LAW JOURNAL [Vol. 70:345

We conclude... that the restrictive agreements standing alone cannot be regarded as violative of any rights guaranteed to petitioners by the Four- teenth Amendment. So long as the purposes of those agreements are effec- hated by voluntary adherence to their terms, it would appear clear that of the agreement there has been no action3 4by the State and the provisions have not been violated. So long as it takes no affirmative action to implement or facilitate discrimination a state may tolerate its existence. Neither its legislature nor its courts have acted to declare it illegal; neither have they directly promoted the cause of discrimination. The state policy of noninterference is not fourteenth amendment state action. Thus, if Shelley's grantor had declined to sell in accordance with the restrictive covenant, and if Shelley had asked a local court to compel the grantor not to comply with that covenant, the court's refusal would have constituted noninterference and would not have been an expression of affirmative state policy for fourteenth amendment purposes. Shelley's situation would then have been the same as Oliphant's. He too would have been unable to obtain a state injunction against "voluntary adherence" to the terms of a discriminatory agreement. What needs to be stressed again is that the difference between Shelley v. Kraenier in fact and Shelley v. Kraemer, or Oliphant v. The Brotherhood, as hypothesized is not that in one case the state, by ordering adherence to a discriminatory agreement, manifests a policy, while in the other it manifests no policy by tolerating a discriminatory agreement. The difference is between the two policies. State action does not extend to a state's "hands-off" policy-a policy which simply tolerates, without affirmatively encouraging, the discriminatory practices of individuals or groups. On this basis, then, Oliphant would seem to be out of luck on the theory of Shelley v. Kraemer. Yet the Shelley theory seems to have gone beyond that case itself and to have aided a litigant who sought to hold a union to the Con- stitution. Perhaps the subsequent development of the Shelley doctrine could provide the springboard for Oliphant's relief.

II Railway Employees' v. Hanson 35 involved the 1951 union shop provision of the Railway Labor Act. The union shop section of the railroad statute is similar to that provided for industry in interstate commerce generally in the Taft- Hartley Act. In the Railway Labor Act, however, union shops are permitted "[n]otwithstanding any ... law . . . of any State." Carrier and union may negotiate an agreement, requiring as a condition of continued employment, that within sixty days following the beginning of [their] employment . . . all employees shall become members of the labor organization . . . Provided, that no such agreement shall require such condition of employment with respect to em- 34. Id. at 13. 35. 351 U.S. 225 (1956). 1961] GOVERNMENTAL ACTION

ployees to whom membership is not available upon the same terms... as are generally applicable to any other member or with respect to employees to whom membership was denied or terminated for any reason other than the failure of the employee to tender the periodic dues, initiation fees, and assessments (not including fines and penalties) 6uniformly required as a condition of acquiring or retaining membership. 3

By his own choice Hanson was not a member of the union which had such an agreement with his employer. In a Nebraska court he sought an injunction against his discharge pursuant to the agreement. His ultimate substantive arguments were that the union shop agreement contravenes the first and fifth amendments of the United States Constitution: that it interferes with "the right to work which the court has frequently included in the concept of 'liberty' within the meaning of the due process clauses," 37 and that it "forces men into ideological and political associations which violate their right to freedom of conscience, freedom of association, and freedom of thought protected by the Bill of Rights." 38 Hanson was unable to prove these allegations in his own case, and lost for failure to substantiate his charges. The Court held "that the requirement for financial support of the collective-bargaining agency by all who receive the benefits of its work is within the power of Congress under the commerce clause and does not violate either the First or Fifth Amendments. '30 However, the Court took care to advise that if a record could be made to show that "the exaction of dues, initiation fees, or assessments is used as a cover for forcing ideological conformity or other action in contravention of the first amendment, this judgment will not prejudice the decision in that case" 40 and that if union assessments "are in fact imposed for purposes not germane to '4 1 collective bargaining, a different problem would be presented." In Machinists v. Street,42 a case from Georgia now pending before the Court, an attempt is being made to provide an adequate factual basis for the complaint.

36. Railway Labor Act, § 2 Eleventh, added by, 64 Stat. 1238 (1951), 45 U.S.C. § 152 Eleventh (1958). 37. 351 U.S. at 234. 38. Id. at 236. 39. Id. at 238. 40. Ibid. 41. Id. at 235. 42. Rcev'd dismissal of petition sub nom. Looper v. Georgia So. & Fla. Ry., 213 Ga. 279, 99 S.E.2d 101 (1957); International Ass'n of Machinists v. Street, 215 Ga. 27, 108 S.E2d 796, probable jurisdictionnoted, 361 U.S. 807 (1959), case set down for rearguinent because Attorney General of the United States not notified that constitutionality of a federal statute in issue, 363 U.S. 825 (1960). The Georgia court held that the "exaction and use" of plaintiffs' money for various political purposes, and the union shop agreements authorizing this action violated the first, fifth, ninth, and tenth amendments. It enjoined the enforcement of the agreements and declared unconstitutional the union shop section-Section 2, Eleventh-of the Railway Labor Act. This decree seems improper. Even assuming that union conduct is to be tested by the Constitution, and that the union has violated the Constitution, there is no reason to con- THE YALE LAW JOURNAL [ Vol. 70 :345

-An intriguing question arises from Hanson: How did the plaintiff succeed in persuading the Court to say that it would consider, and in limited fashion to consider, the substantive first and fifth amendment questions? On the strength of Shelley it would appear that Hanson, like Oliphant, would have failed, absent any federal or state legislation, had he argued that the refusal of a state court to enjoin enforcement of a union shop agreement raised fourteenth amendment questions. He would have failed since, like Oliphant, he could not have shown sufficient state involvement to come within the state action requirements articulated in Shelley v. Kraemer. Moreover, the result would in principle be the same if the state had a statute -and it was the only piece of labor legislation in the state-providing that unions and employers were free, if they wished, to negotiate union shop agreements. It makes no difference for fourteenth amendment purposes whether the state's "hands-off" policy is enunciated by court or legislature. Further, if the action were brought under a comparable federal enactment no governmental action would be found. It would seem to follow, in addition, that if we had a "hands-off" state policy and a permissive federal statute combined in one case, no substantive constitutional question would be presented, for no governmental action-state or federal-could be shown. But Hanson arose in Nebraska which by constitution and statute forbids the union shop.43 Mr. Justice Douglas, speaking for the

strue section 2, Eleventh as empowering the union to act unconstitutionally. Indeed, nor- mal canons of construction would lead to a contrary reading of the statute. Cf. United States v. CIO, 335 U.S. 106 (1948). Furthermore, it would seem, as the brief for the Government points out, that the "courts below also committed error when they sweeping- ly declared on this record the union shop agreements to be 'null, void, and of no effect as between the parties.' ... Hanson held that these agreements are not null and void, but are valid in general and on their face. The illegality of the particular expenditures would not invalidate the whole agreement." Brief for the United States, pp. 39-40, International Ass'n of Machinists v. Street, No. 4, Oct. Term 1960, Supreme Court of the United States. In its brief, the Government has also taken the position that "the Court need not and should not determine the constitutionality or legality of the various expenditures and activities which appellees challenge." Id. at 49. The Government explains its position this way: The courts below and the parties in this court have not attempted to distinguish and differentiate among the many kinds of union activities and expenditures involved. Appellants have argued that all of their activities or expenditures are proper, while appellees contend that all expenditures not directly related to the negotiation, maintenance and administration of collective bargaining agreements violate their constitutional rights. The various activities and expenditures in dispute have been lumped together, both in the record and in the contentions of the parties. Only in a proceeding to enjoin the'use of the funds-received from appellees'-fees and dues for each kind of disputed expenditure (or comparable litigation)--either on remand or in a new action-would the factual background and legal considerations concerning each class emerge clearly. And only in such a proceeding, we submit, should this Court undertake the decision of the constitutional questions tendered in this case. Id. at 51-52. 43. NEB. CoNsT. art. XV, § 13; NEB. REv. STAT. § 48-217 (1952). 19ki] GOVERNMENTAL ACTION

Court, indeed went far to make the governmental action issue turn on this difference. The union shop provision of the Railway Labor Act is only permissive. Congress has not compelled nor required carriers and employees to enter into union shop agreements. The Supreme Court of Nebraska neverthe- less took the view that justiciable questions under the First and Fifth Amendments were presented since Congress, by the union shop provision of the Railway Labor Act, sought to strike down inconsistent laws in 17 states .... The Supreme Court of Nebraska said, "Such action on the part of Congress is a necessary part of every union shop contract entered into on the railroads as far as these 17 States are concerned for without it such contracts could not be enforced therein." 160 Neb., at 698, 71 N.W. 2d, at 547. We agree with that view. If private rights are being invaded, it is by force of an agreement made pursuant to federal law which expressly declares that state law is superseded. Cf. Smith v. Allwright, 321 U.S. 649, 663. In other words, the federal statute is the source of the power and authority by which any private rights are lost or sacrificed. Cf. Steele v. Louisville & N.R. Co., 323 U.S. 192, 198, 199, 204... ; Rail- road Trainmen v. Howard, 343 U.S. 768; Public Utilities Comrn'n v. Pollak, 343 U.S. 451, 462. The enactment of the federal statute authorizing union shop agreements is the governmental action on which the Con- stitution operates, though it takes a private agreement to invoke the federal sanction. As already noted, the 1951 amendment, permitting the negotiation of union shop agreements, expressly allows those agreements notwithstand- ing any law "of any State".. . . A union agreement made pursuant to the Railway Labor Act has, therefore, the imprimatur of the federal law upon it and, by force of the Supremacy Clause of Article VI of the Constitution, could not be made illegal nor vitiated by any provision 44 of the laws of a State.

This may be a logical extension of Shelley v. Kraemer. The parties to the restrictive covenant in that case could not have accomplished their purposes if the state had not ordered compliance with their agreement. This affirmative act of the state was sufficient for a Supreme Court finding of governmental action. In Shelley, state law did more than simply tolerate the private agreement. In Hanson, federal law did more than simply tolerate the union shop agreement; it made it possible by the ousting of inconsistent state law. Absent the Railway Labor Act, Nebraska law would prevent the union and the carrier from concluding a union shop agreement. The permissive federal law which ousts the prohibiting state law is necessary, therefore, if the parties are to enter into their contract at all and are to accomplish the purposes provided for in their union shop agreement. The, partnership betweeft uniof and government which is requisite to a finding of governmental action thus is made out. If the Court means what it seems to say, however, Hanson could lead to in- congruous results. In any given case the Court would have to look to state law to decide whether there was sufficient federal involvement to hold the private 44. 351 U.S. at 231-32. THE YALE LAW JOURNAL [Vol. 70:345 agreement to first and fifth amendment standards. Where state law allowed the union shop, the parties could have achieved their purposes without the federal enactment. Where this is the case the federal statute in fact does no more than declare that the federal government will "tolerate" the union shop on the railroads. And governmental toleration will not sustain a finding of governmental action. Judge Learned Hand reached a conclusion very like this in Otten v. Baltimore & O.R.R.,45 a Railway Labor Act case, involving an employee whose religious beliefs would not allow him to join a union. The case is similar to Hanson, but it arose in New York where state law tolerates the union shop.46 Therefore the Constitution had no application. But if Otten had arisen in Nebraska, where state law forbids the union shop, the Constitution's prohibition against discriminatory governmental action might have protected the unfortunate employee who declined on religious grounds to join the union, for in Nebraska, the union shop would not have been possible in the absence of federal law. This unhappy result suggests that the Court did not mean all it said in Hanson. A further reason for doubting that the Court intended its opinion to be taken literally is that the Hanson reasoning would then conflict with the major justification for pre-emption in labor-management matters-uni- form rules throughout the nation.47 Perhaps what the Hanson Court did mean is that state law, as such, is irrelevant in determining whether there has been governmental action. It is a matter of indifference what the state law was before the federal statute was passed, because the state is now no longer free by changing its laws to grant rights to someone like Hanson through the enactment of a "Right to Work" statute. If this is what the Court was saying, however, the concept of governmental action has indeed become attenuated. When an ousted state law is consistent with federal law, the impact of the federal statute is, in fact, zero. Only after a conceptual somersault can the federal enactment be said to give rise to governmental action. There is, moreover, an additional difficulty with the governmental-action- arising-from-pre-emption interpretation of Hanson-that governmental action automatically results when a congressional enactment nullifies, or even con- firms existing state law. It gives to federal governmental action under the fifth amendment a meaning quite different from the one the Court has given to state action under the fourteenth. It means that unlike the states, the federal government in many situations may not explicitly adopt a "hands-off" policy. It 45. 205 F2d 58 (2d Cir. 1953), aff'd sub nor. Otten v. Staten Island Rapid Transit Ry., 229 F.2d 919 (?d Cir.), cert. denied, 351 U.S. 983 (1956) ; accord, Wicks v. Southern Pac. Co., 231 F.2d 130 (9th Cir.), cert. denied, 351 U.S. 946 (1956). 46. "[Mn New York, where the plaintiff was employed and where the agreement im- pinged upon him, a 'union shop,' and perhaps even a 'closed shop,' agreement was valid at the common law .... National Protective Association v. Cumming, 170 N.Y. 315, 63 N.E. 369, . . . Jacobs v. Cohen, 183 N.Y. 207, 76 N.E. 5.... Williams v. Quill, 277 N.Y. 1, 12 N.E.2d 547." Otten v. Baltimore & O.R-R., 205 F.2d 58, 60 & n.5 (2d Cir. 1953). 47. See, e.g., San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236 (1959) ; Well- ington, Labor and the Federal System, 26 U. CHi. L. REv. 542 -(1959). 1961] GOVERNMENTAL ACTION

means that all private action taken under the authority of federal legislation that occupies a field by that token alone becomes governmental action. There- fore, the federal government must affirmatively outlaw private behavior that conflicts with the standards applicable to it under the fifth amendment. It is surely not self-evident that such a difference between federal and state governmental action-a difference which may or may not be desirable-is a by- product of the supremacy clause and the doctrine of pre-emption. Those instruments of policy were fashioned to cope with problems unrelated to the question of when constitutional restraints apply to the actions of private groups. One must conclude that if the Court had meant to use the supremacy clause and pre-emption to extend the concept of governmental action in this novel way it should have given explicit consideration to the problems involved. Two theories other than pre-emption, however, may more adequately justify the finding of governmental action in Hanson. The citation in the Court's opinion to Smith v. A11wright 48 suggests an analogy to the white primary cases. In those cases, despite successive attempts by the state of Texas to disengage itself from the Democratic primary (in order to maintain the pure white character of those all-important elections), the Supreme Court found unconstitutional state action.49 The cases seem to in- dicate that once a government has actively participated in an activity, it is much easier for the Court to continue to find state action after the state attempts to remove itself from that activity than it would have been in the absence of any earlier state involvement.5 0 Yet, the Railway Labor Act, which from 1934 to 1951 prohibited union shop agreements,51 cannot accurately be considered within the Smith v. Allwright context. For while it is true that prior to 1951 federal protection was extended to employees who along with Hanson did not wish to join a union, the All- wright and Hanson situations are significantly distinguished by the differing reasons for "governmental disengagement" in each. In the one case Texas clearly evinced its desire to perpetuate a discriminatory practice which it earlier had sponsored. In Hanson, on the other hand, the Court tells us that a wholly different objective underlay the Railway Labor Act amendment. Those provisions [prohibiting union shop agreements] were enacted in 1934 when the union shop was being used by employers to establish and

48. 321 U.S. 649 (1944). 49. Nixon v. Herndon, 273 U.S. 536 (1927); Nixon v. Condon, 286 U.S. 73 (1932); Grovey v. Townsend, 295 U.S. 45 (1935); cf. United States v. Classic, 313 U.S. 299 (1941) ; Smith v. Allwright, 321 U.S. 649 (1944) ; Terry v. Adams, 345 U.S. 461 (1953). 50. Cf. Pollak, Racial Discrimination and Judicial Integrity: A Reply to Professor WVcchslcr, 108 U. PA. L. REv. 1, 19-22 (1959). 51. "The union shop provision of the Railway Labor Act was written into law in 1951. Prior to that date, the Railway Labor Act prohibited Union Shop agreements. 48 Stat. 1186, 45 U.S.C. § 152 Fourth and Fifth; 40 Op. Atty. Gen. 254." Railway Employees' v. Hanson, 351 U.S. 225, 231 (1956). THE YALE LAW JOURNAL [Vol. 70 :345

maintain company unions, 'thus effectively depriving a substantial number of employees of their right to bargain collectively.'2 . . . By 1950 company unions in this field had practically disappeared.5

The difference then is "disengagement" for the purpose of saving a state-in- stituted scheme of discrimination as contrasted with "disengagement" for a constitutionally inoffensive purpose. Reliance on the white primary cases as a rationale for the Hanson decision may be misplaced for an additional reason. Those cases involve the fifteenth amendment, and it has been argued by Professor Pollak that there is little transferability between the law under that amendment and the usual notions concerning governmental action under the fourteenth. Yet how can the fifteenth amendment apply where the fourteenth does not, since both are addressed to state action? The question generates its own answer: with respect to the particular problem to which the fifteenth amendment is addressed-protecting the right of Negroes not to be dis- criminated against at the polls-the amendment must impose on the states a heavier affirmative duty to assure an equal franchise than does the fourteenth. If this were not so, the fifteenth amendment would be a redun- dancy, having no scope for separate and effective application. Construing the fifteenth amendment to be an independent meaningful guarantee is in harmony with its framers' comprehensive purposes. "It was . . . well understood in Congress at the time the amendment was under consideration that it applied to any election, from that of a presidential election down to the most petty election for a justice of the peace or a fence-viewer." . . . But the Court had learned in Classic that in the deep South the only opportunity to exercise the constitutionally protected right to vote arises at the primary... balloting. Thus to find no state duty to prevent the exclusion of Negroes from the only elections which matter would be to delete the fifteenth amendment from the Constitution.r 3

A more promising explanation of the Hanson case arises from the fact that the union shop provision of the Railway Labor Act is but one section of a comprehensive regulatory statute. This statute establishes a federal program for the railroads designed to encourage collective bargaining.M And collective bargaining "generally has been considered to absorb and give statutory approval to the philosophy of bargaining as worked out in the labor movement in the

52. Ibid. 53. Pollak, supra note 50, at 22-23. Professor Pollak goes on to say: Imposing fourteenth amendment restraints on a southern Democratic primary would have to be predicated on a judgmefnt that managing the electoral process is an inalienable sovereign functi6n, and that whoever does that managing acts on the 'state's behalf .... But the argument, whatever its validity, is one which goes far beyond the limited guarantee of racial equality in the political process embodied in the fifteenth amendment and properly vindicated in Smith v. Allwright and Tenry v. Adams. Id. at 23. 54. 44 Stat. 577 (1926), as amended, 48 Stat. 1185 (1934), 49 Stat. 1189 (1936), 54 Stat. 785, 786 (1940), 64 Stat. 1238 (1951), 45 U.S.C. §§ 151-88 (1958). 1961] GOVERNMENTAL ACTION

United States." 55 This is a philosophy in which union security is a central 0 tenet." Thus, while the act's union shop provision taken alone is quite properly read to embody a federal policy of toleration, taken as part of an organic whole-read in context, as statutes must be read-the union shop provision embodies a federal policy which encourages the formation of union security agreements. It is perhaps this encouragement that provides sufficient governmental action to make the first and fifth amendments applicable. Can this explanation of Hanson be developed into a theory in aid of Lee Oliphant? How tenuous is the extension needed to establish doctrinally the sought after governmental action in his case? A brief examination of existing federal labor law suggests it is not tenuous at all. Federally sponsored collective bargaining is central to labor-management relations in this country. The Railway Labor Act and the National Labor Relations Act protect unions in their organizing campaigns from a variety of time-tested, employer-directed, obstructionist techniques. 57 These statutes require the employer to bargain in good faith with the union that is designated as representative by a majority of the employees in an appropriate (often gov- ernmentally determined) unit. 8 By federal law the chosen union represents all employees-members and nonmembers, employees who voted for, and employees who voted against, the union-in the negotiation and administration of the collective agreement. 9 By federal law, too, this union is the exclusive employee bargainer. 60 Whether he likes it or not, the employee who did not choose the majority union is generally bound by its acts in drafting and applying the code which regulates his industrial life.0 1 The individual employee can participate meaningfully in this vital process only through the union; and membership is the condition precedent to such participation. By sponsoring collective bargaining and the institutions associated with it as part of our national labor policy the federal government has helped to make union membership important to the working man. Conceptually, at least, is this not sufficient involvement for purposes of governmental action under the fifth amendment? Is there not, indeed, much the same degree of

55. Order of R.R. Telegraphers v. Railway Express Agency, Inc., 321 U.S. 342, 346 (1944); see J. I. Case Co. v. NLRB, 321 U.S. 332 (1944). 56. See Golden & Ruttenberg, The Union Shop is Democratic and Necessary, in BAKKE & KERR, UNIONS, MANAGEM2,ET AND THE PUBLIC 129 (1948). 57. National Labor Relations Act § 8(a), 61 Stat. 140 (1947), as amended, 29 U.S.C. § 158(a) (1958) ; Railway Labor Act § 2, 44 Stat. 577 (1926), as amended, 45 U.S.C. § 152 (1958). 58. National Labor Relations Act §§ 8(a) (5), 9, 61 Stat. 140 (1947), as amended, 29 U.S.C. §§ 158(a) (5), 9 (1958) ; Railway Labor Act § 2 Fourth, 44 Stat. 577 (1926), amended, 45 U.S.C. § 152 Fourth (1958). 5t. Note 58 supra; see J. I. Case Co. v. NLRB, 321 U.S. 332 (1944) ; Order of R.R. Telegraphers v. Railway Express Agency, Inc., 321 U.S. 342 (1944). 60. Note 58 supra. 61. Note 58 supra. THE YALE LAW JOURNAL [Vol. 70: 345

federal participation which it was earlier suggested might explain the Hanson case? Perhaps this easy extension of doctrine can be carried even further. The extent to which federal statutes have given unions control over the economic well-being of the bulk of employees in an enterprise suggests the delegation to the union of governmental power. For purposes of finding the Constitution applicable, the situation may not be unlike that of the company town in Marsh v. Alabama.62 While the town was owned by a single private corporation, it had all the attributes of an ordinary municipality. The Supreme Court held, therefore, that its affairs had to be conducted in accordance with the dictates of the fourteenth amendment. The suggested explanation of Hanson, and the easy extrapolation of this explanation to cover Oliphant, is not without difficulties. Once embraced, how far is it likely to push the concept of governmental action? For example, if the extent of government participation is crucial, can Hanson and Oliphant be distinguished doctrinally from the very different problems of the charitable trust? Professor Clark has stated: The power to dispose of property at death is a privilege granted by law and supervised through probate and administration by courts and judicial- ly appointed fiduciaries.... The trust becomes operative only after a court has found, either specifically or by inference, that it is charitable. Nor has government remained neutral. To encourage a continuous flow of funds into philanthropic enterprises, it bestows privileges of which tax immunity is only one. The state creates and defines charitable trusts, grants them perpetual existence, modernizes them through cypres, appoints and regulates the trustees, approves accounts, construes ambiguous language in the trust charter and sometimes goes so far as to impose a less stringent standard of tort liability on such trusts than on their private counterparts. 3 What emerges from this discussion, then, is that only a slight doctrinal extension from existing case law is needed to bring the Constitution to bear on many kinds of activities now considered private. Clearly, it is all that is needed doctrinally to bring the Constitution to bear on union conduct. But, of course, as every thinking lawyer knows, conceptual consistency, while always desirable and often important, is itself insufficient to justify the extension of existing doctrines to new problems. We may acknowledge that in Hanson, in its sequel Street, and in Oliphant, there are problems in need of solution; that by a slender extension of existing doctrine the Constitution could be used in an attempt to solve these problems. But how far, if at all, doctrine should be extended in a particular area depends in large part upon the consequences of its extension in any given case, and upon the alternative methods available for solving the problems in question. It is to these matters that we now turn.

62. 326 U.S. 501 (1946). 63. Clark, Charitable Trusts, the Fourteenth Amendment, and the Will of Stephen Girard, 66 YALE L.J. 979, 1003-04 (1957). 1961] GOVERNMENTAL ACTION

CONSEQUENCES AND ALTERNATIVES Responsibility and power over federal labor policy shifts from Congress to the Court when union conduct is judged by the Constitution. This is not to say that Congress ever relinquishes the role of initiator-in-chief of such policy. It does not. Nor is the Court's principal law-making role of interpreter of congressional purpose affected. But when union conduct is judged by constitutional standards Congress must share, and the Court must assume, responsibility for fashioning afresh federal policy. This is only to say that the Court takes on a new role. It has the responsibility for deciding without direct congressional guidance whether unregulated union conduct squares with the Constitution. For this purpose, and in this sense, the Court becomes the initiator of federal labor policy. I How well suited is the Court to assume this new role in the cases which have engaged our attention? Perhaps some light may be shed on this issue by examining a situation where courts, in the guise of statutory interpretation, have assumed a similar role. The Supreme Court has read both the Railway Labor Act and the National Labor Relations Act to impose a duty of fair representation on the collective bargaining agent." The union must represent fairly all the employees it serves in the negotiation and administration of the collective agreement.0 5 Congress has made no efforts to define the specific content of that duty, or to supply the standards with which courts may judge union conduct which has allegedly transgressed the statutory duty. Substantial doubt exists as to whether the doctrine of fair representation in its present form adequately protects employees. 0 The problem is especially acute in the administration of collective agreements through grievance procedures. It is often extremely difficult to ascertain whether a union satisfactorily and disinterestedly balanced the competing claims of employee groups before it assumed a position with management on the disposition of a particular grievance or group of grievances. 7 This is subtle business. Yet many cases have tested a union's compliance with its duty of fair representation. 68 While litigation in this area does not directly involve the appli-

64. Steele v. Louisville & N.R.R., 323 U.S. 192 (1944); Syres v. Oil Workers Int'l, Local 23, 350 U.S. 892, reversing per curiam, 223 F.2d 739 (5th Cir. 1955) ; Wallace Corp. v. NLRB, 323 U.S. 248, 255-56 (1944) ; Ford Motor Co. v. Huffman, 345 U.S. 330, 337-39 (1953). 65. Ibid.; Conley v. Gibson, 355 U.S. 41 (1957). 66. See Wellington, Union Democracy and Fair Representation: Federal Responsi- bility in a Federal System, 67 YALE L.J. 1327, 1339-43 (1958). 67. Id. at 1336; see Cox, The Duty of Fair Representation, 2 VILL. L. REV. 151 (1957); Note, Duty of Union to Minority Groups in the Bargaining Unit, 65 HARv, L. RaV. 490 (1952). 68. See, e.g., cases cited notes 64-65 supra; Brotherhood of R.R. Trainmen v. Howard, 343 U.S. 768 (1952); Colbert V. Brotherhood of R.L Trainmen, 206 F2d.9 (9th Cir. THE YALE LAW JOURNAL [Vol. 70: 345

cation of the Constitution to union conduct (the duty is statutory), the judi- cially fashioned standard is derived from constitutional law. In the leading Steele case,609 where the Brotherhood of Locomotive Firemen and Enginemen had negotiated an agreement making seniority rights turn upon race, the Court said: We think that the Railway Labor Act imposes upon the statutory representative of a craft at least as exacting a duty to protect equally the interests of the members of the craft as the Constitution imposes upon a legislature7 0 to give equal protection to the interests of those for whom it legislates. This test is easily applied where a union draws distinctions among employees because of race. It has little utility in nonracial situations. Where the legitimacy of economic distinctions is at stake the equal protection clause is a crude instrument. Though at the outer limits it can separate what is reasonable from what is unreasonable,71 that affords an economic minority in the bargaining unit small protection indeed. For example, for a union at the time of the merger of two companies to insist that all employees in the larger of the two establish- ments be given a higher seniority rating than any employee in the smaller establishment may be contrary to usual practice and to the expectations of the employee community. Dovetailing may be more in keeping with both custom and expectations. But the union's solution is certainly a reasonable one, if rea- 72 sonable is given its constitutional-law meaning. As a practical matter, however, courts must use a relatively loose standard like "reasonableness" when judging union conduct. The alternative is for courts to assume a task they probably could not perform well. Problems vary from industry to industry and from union to union. The internal structure and politics of a union, the history of collective bargaining and the present power relations reflected in it, and the state of development and economic condition of an industry, are the obvious, though by no means the exclusive, interacting factors which affect judgments about competing employee claims. A court is institutionally unable to second guess union judgments which result from a weighing and balancing of such dynamic variables. And as one moves away from tests of reasonableness one moves ineluctably into the second guessing game. Pos-

1953), cert. denied, 346 U.S. 931 (1954) ; Lewellyn v. Fleming, 154 F.2d 211 (10th Cir.), cert. denied, 329 U.S. 715 (1946). 69. Steele v. Louisville & N.R.R., 323 U.S. 192 (1944). 70. Id. at 202. 71. See Morey v. Doud, 354 U.S. 457, 463-64 (1957) ; Quaker City Cab Co. v. Penn- sylvania, 277 U.S. 389, 406 (1928) (Brandeis, J., dissenting). "[T]he equality clause requires merely that the classification shall be reasonable. We call that action reasonable which an informed, intelligent, just-minded, civilized man could rationally favor."). 72. See Trailer Co. of America, 51 N.L.R.B. 1106 (1943); Trailer Co. of America, 53 N.L.R.B. 1248 (1943) ; Hess v. Trailer Co. of America, 31 Ohio Op. 566 (C.P. 1944) ; Whirls v. Trailmobile Co., 64 F. Supp. 713 (S.D. Ohio 1945), aff'd, 154 F2d 866 (6th Cir. 1946), rev'd, 331 U.S. 40 (1947) ; Britt v. Trailmobile Co., 179 F2d 569 (6th Cir.), cert. denied, 340 U.S. 820 (1950). See generally Kahn, Seniority Problems in Business Mergers, 8 IN . & LAB. Ru. REv. 361 (1955). 1961] GO VERNMENTAL ACTION sibly an expert tribunal could do the job, but such an agency is far removed indeed from a federal district court.73 In the fair representation cases, then, courts have talked the language of constitutional law, and have acted as they would have to if they were in fact applying the Constitution. They have therefore in nonracial situations usually found for the union.74 This digression into the problems of fair representatiQn-and it is admitted- ly a digression, for the problems are not constitutional-is apposite since the institutional difficulties here seem related to those the Court will face if it should undertake to apply the Constitution to union conduct in Hanson's sequel, Ala- 75 chinists v. Street. It will be recalled that the Court in Hanson advised future litigants, of whom Street is one, that if under a union shop assessments "are in fact imposed for purposes not germane to collective bargaining, a different problem would be presented." 76 This dictum seems to mean that the fifth amendment may be violated when union funds are not used for collective bargaining purposes.77 judicial review of union action here potentially involves the Court in second guessing the union. It must decide whether money spent in support of proposed state or federal legislation, or for the election of a particular candidate for public office is or is not germane to the union's role as bargaining agent. Re- view, therefore, takes the conscientious Justice deep into the history, purpose, and ideals of the union movement in general, and of the union involved in particular. It takes the Justice deep into the collective bargaining relationships of that union, and this in turn necessitates that he explore the history of the industry with which the union negotiates, its attitudes, and its structure.78 Realistically, the process invites the establishment of one of two standards: a "reasonableness" test, which will work out the way the doctrine of fair representation has worked out, and which comes close to giving the union a carte

73. My views upon this are developed at some length in Wellington, supra note 66, at 1357-61. 74. See, e.g., Ford Motor Co. v. Huffman, 345 U.S. 330 (1953) ; Jennings v. Jennings, 91 N.E.2d 899 (Ohio Ct. App. 1949). "It is significant that in cases where challenged changes in seniority provisions did not involve racial discrimination, the courts consistently have refused to upset the agreement." Pellicer v. Brotherhood of Ry. & S.S. Clerks, 118 F. Supp. 254, 258 (S.D. Fla. 1953), aff'd, 217 F.2d 205 (5th Cir. 1954), cert. denied, 349 U.S. 912 (1955). One can only speculate about the impact of this on Congress, but it seems entirely possible that the quite proper judicial behavior in this area has tended to bury problems of the minority employee which may exist. Since court performance here is much like a determination by the Court that congressional or state action is constitutional, this should not be surprising. There surely is a tendency on the part of the vast uninitiated to equate that which is constitutional to that which is good. 75. Supra note 42. 76. 351 U.S. at 235. 77. Id. at 232-33. 78. Compare the judicial experience with § 304 of the LMRA, recounted in SMITH & MERRIFIELD, LABOR RELATIONS LAW 1154-63 (1960). THE YALE LAW JOURNAL [Vol. 70: 345 blanche; or a test which translates into constitutional law the private philosophy of the Justices. The former accomplishes little, except to surround conduct which perhaps should be regulated by Congress with a halo of constitutionality,O while the latter, as history reveals, is intolerable.8 0

II In Hanson the Court also advised potential litigants that if they could demonstrate that "the exaction of dues, initiation fees, or assessments is used as a cover for forcing ideological conformity or other action in contravention of the first amendment, this judgment will not prejudice the decision in that case."31 In Street the plaintiffs allege, inter alia, and the Georgia court has found as a fact, that "the union dues and other payments [plaintiffs] ... will be required to make to the union will be used to 'support ideological and political doctrines and candidates' which [plaintiffs] . . . are unwilling to support and in which they do not believe."8 2 If the Court, within the analytical framework of union activity as governmental action, is to decide whether this use of funds by the union contravenes the first amendment, it would seem that the Justices must consider several factors: on the one hand are to be weighed the uses and purposes to which the money is to be put, the importance of the objectives in question to the labor organization, and the extent to which they are supported by the majority within the organization, and on the other hand there is to be assessed the impact of the union's action on the dissenting employee. As noted, this means the Court's immersion in the history, structure, and aspirations of the union movement, and of the particular union. In short, it means immersion in collective bargaining, and an understanding of the relationship between economic power and political action. A national policy of collective bargaining requires a rough equivalence of economic power between labor and management. The economic position of both labor and management-their power at the bargaining table-is dependent upon many variables, not the least of which is ever changing federal and state law. The impact upon economic power of federal legislation which makes certain employer and union practices illegal is obvious. A union, for example, may not apply secondary pressures to bring its adversary to terms.8 3 And its freedom to engage in organizational picketing is limited. 4 Less obvious, but also important to the power of a union at the bargaining table, are minimum wage

79. See BLAcK, THE PEOPLE AND THE COURT (1960). 80. See, e.g., Adair v. United States, 208 U.S. 161 (1908). 81. 351 U.S. at 238. 82. 213 Ga. at 284, 99 S.E.2d at 104. 83. Labor-Management Relations Act § 8(b) (4), 61 Stat. 141 (1947) (now § 8(b) (4), as amended, Labor-Management Reporting and Disclosure Act § 704(a), 73 Stat, 542 (1959), 29 U.S.C. § 158(b) (4) (Supp. 1959). 84. The Labor-Management Reporting and Disclosure Act of 1959, § 704(c), 73 Stat. 544 (1959), 29 U.S.C.A. § 158(b) (7) (Supp. 1959), added a new union unfair labor practice,'§ 8(b) (7), to the National Labor Relations Act § 8, 49 Stat. 452 (1935), as amended, Labor-Management Relations Act, 61 Stat. 141 (1947), 29 U.S.C. § 158 (1958). 1961] GOVERNMENTAL ACTION legislation,85 social security legislation,86 legislation dealing with unemploy- 8 ment 8 and workmen's compensation," and the many other forms of welfare legislation which provide a foundation upon which unions may build in bargaining with management. 89 Another factor that may be equally important to the union's economic position at the bargaining table is tariff legislation or other types of industry protecting or subsidizing enactments.90 More attenuated perhaps, but still important, are the general economic policies of an Adminis- tration. Is it then any wonder that business-minded unions are interested in politics and politicians? 91 Thus the task of the court is complicated. And little can be gained from past judicial experience in determining the importance of any particular union-supported legislative program, and in balancing it with the impact of the union's action on the dissenting employee. Each case will present the Court with a dis- crete problem. And even with respect to any one union the problem will change from time to time with shifts in union economic power-shifts caused by such

85. See generally for a textual and case study of governmental wages and hours legislation, LABOR LAw GRoup TRUST, THE EMPLOYIMENT RELATION AND THE LAW 395- 527 (Aaron, ed. 1957). 86. Id. at 528-813. 87. Ibid. 88. Id. at 96-382. 89. The brief for the Appellants in the Street case suggests the type of welfare legislation a union which bargains with a railroad carrier is likely to be interested in. The record is replete with instances of the type of legislation in which labor organizations interest themselves, and in specific legislation in which some of the defendant labor organizations interested themselves .... First and foremost, are the Railroad Retirement and Unemployment Insurance Acts, the railroad equivalents of state unemployment insurance laws and the Social Security Act .... Another is the Federal Employers' Liability Act. A third type of legislative activity pertains to efforts to equalize the competitive positions of railroads with other forms of trans- portation, thus protecting the work of railroad employees. Another is legislation directed to safety, to reduce the physical hazards of railroad employment. Other legislative activity in which it engages seeks to protect the purchasing power of wage earners. There, are three model bills which this Brotherhood seeks to have enacted in all the States, and has had thus far some small measure of success. The first would prohibit employers that require medical examinations of their employees, a common practice on the railroads, from charging the employees for such examinations. The second would require employers that pay their employees by check to make arrange- ments under which the employees could cash those checks without being charged a fee. The third would establish certain standards of health and sanitation in working conditions. Brief for Appellants, pp. 59-60, International Ass'n of Machinists v. Street, No. 4, Oct. Term 1960, Supreme Court of the United States. 90. For example, other things being equal, the unions which bargain with American watch manufacturers are quite likely to be in a better position at the bargaining table when there is a high, rather than when there is no, tariff on Swiss watches. 91. See, e.g., REYNOLDS, LABOR EcoNOMICS AND LABOR RELATIONS 80-81 (2d ed. 1958). THE YALE LAW JOURNAL [ Vol. 70: 345 factors as changed economic conditions or changed federal and state law.v2 Furthermore, because unions are very different institutions indeed from States, municipalities or the federal government for that matter, it will be hard for the Court to transfer the wisdom contained in traditional first amendment decisions to its review of union conduct.93 III

Why should the Court allow itself to be forced into these exercises-an in- evitable consequence of extending the Constitution to cover labor union activity -when another arm of government, Congress, is more capable institutionally of dealing with these problems? This is the ultimate question. It asks, in effect, whether there is reason enough in this area for judicial review. Few students of American Government now question the need for judicial review of the constitutionality of congressional enactments. Controversy exists over the nature of review, not over its legitimacy.94 The basis for the wide-

92. This is just to say that the importance to a union of political activity as an alternative to economic action depends in part upon the economic strength of the union, and this in turn, is affected by the state of the law surrounding collective bargaining. 93. Any resolution of the underlying substantive constitutional question tendered by the Street case may have important implications for the first amendment rights of the labor union itself, and accordingly, for the power of Congress and the states to regulate union political action by means of corrupt practices legislation. The federal Corrupt Practices Act is Section 304 of the Taft-Hartley Act which provides in part that: "It is unlawful for ... any labor organization to make a contribution or expenditure in connection with any election at which Presidential and Vice Presidential electors or a Senator or Representative in, or a Delegate or Resident Commissioner to Congress are to be voted for, or in connection with any primary election or political con- vention or caucus held to select candidates for any of the foregoing offices, or for any candidate, political committee, or other person to accept or receive any contribution prohibited by this Section." The justified fear of government controlled by groups and not by individuals is of course the principal reason for regulating the political activities of groups. On the one hand, regulation can be of the corrupt practices type, aimed at limiting group expenditures for political purposes and thereby increasing the political power of the individual. On the other hand, regulation can be directed towards the problems exposed by the Street and Hanson cases. Regulation of one kind, of course, does not necessarily preclude regulation of the other kind. But the problems are of a piece. If the Court were to determine that the union had not abridged Street's first amendment rights, would not the need to control the union's use of its money for political purposes be intensified? The extent of the need to protect the purity of the political process surely must be considered by the Court in deciding the validity of Section 304 of the Taft- Hartley Act or other similar federal or state legislation restricting the union's freedom of political action. If earlier the Court had limited the power of Congress to restrict the political freedom of unions, might this not in turn affect Street's first amendment claim, assuming the Court were to decide the substantive first amendment question he raises? 94. Cf. Wechsler, Toward Neutral Principles of Constitutional Law, 73 HARV. L. REV. 1 (1959) ; Pollak, supra note 50; Black, The Lawfulness of the Segregation Decisions, 69 YALE L.J. 421 (1960) ; Mueller & Schwartz, The Principle of Neutral Principles,7 U.C. L.A.L. REv. 571 (1960). 19611 GOVERNMENTAL ACTION spread acceptance of judicial review probably has little to do with the inten- tions of the founding fathers. It is pragmatic. The minority must be protected from legislative abuse by the majority. Congress has developed in a fashion- which might or might not have been different absent judicial review-that pre- cludes sole reliance upon congressional self-policing. In much the same manner, few students of American labor today believe that sole reliance upon union self-policing would adequately protect the interests of minority employees represented by unions. The labor union as an institution has not grown up with the degree of responsibility required for such an ideal result. But while only the Court can protect minorities in the community at large from congressional abuse, Congress itself may be quite up to protecting from abuse minority employees within unions. 95 In fact, these minorities in many situations receive adequate representation in Congress. The interests of organized employer groups are frequently advanced by the advocacy of the minority employees' cause. This, coupled in some congress- enexiwith a high sense of legislative responsibility, affords employee minorities legislative protection. The history of the Labor-Management Reporting and Disclosure Act of 1959 provides ample evidence of this fact. When hearings before the McClellan Committee 96 and before the House 0T and Senate Labor Committees Il disclosed that some unions were abusing their positions to the disadvantage of their members, the pressure on Congress to regulate was very great indeed. This pressure soon was translated through the legislative process into an extensive statutory code regulating such internal union affairs as elections,9 trusteeships, 1°° disciplinary proceedings, 10 1 and financial transactions. 0 2 One may suppose that these same pressures will in time be

95. Congress has recently made a major effort in this direction by enacting the Labor- Management Reporting and Disclosure Act of 1959, 73 Stat. 519 (1959), 29 U.S.C. §§ 153, 158, 159, 160, 164, 186, 187, 401, 402, 411-15, 431-40, 461-66, 481-83, 501-04, 521-31 (Supp. 1959). 96. Hearings Before the Select Committee on Improper Activities in the Labor or Management Field (1957-1959) ; see S. REl'. No. 1417, 85th Cong., 2d Sess. (1958). 97. H.R. REP. No. 741, 86th Cong., 1st Sess. (1959). 98. Hearings on Labor-Alanagement Reform Legislation Before the Subcommittee on Labor of the Senate Committee on Labor and Public Welfare, 86th Cong., 1st Sess. (1959). A short history of the birth of the LMRDA may be found in Aaron, The Labor-Manage- ment Reporting and Disclosure Act of 1959, 73 HARv. L. REv. 851, 851-55 (1960). See also Levitan, Union Lobbyists' Contribntions to Tough Labor Legislation, 10 LaB. L.J. 675 (1959). 99. LMRDA, tit. IV. 100. LMRDA, tit. III. 101. LMRDA, tit. I, § 101(a) (5). 102. LIMRDA, tit. II & V. See also LMRDA, tit. I, § 101 (a) (1) which provides: Every member of a labor organization shall have equal rights and privileges within such organization to nominate candidates, to vote in elections or referendums of the labor organization, to attend membership meetings, and to participate in the THE YALE LAW JOURNAL [Vol. 70: 345

effective to protect people like Hanson and Street, if the evidence reveals that theirs is a situation in need of protection. Indeed, Congress has in the past been sensitive to observed problems resulting from union security agreements.'0 3 And when Congress does take action it can make accommodations

deliberations and voting upon the business of such meetings, subject to reasonable rules and regulations in such organization's constitution and by laws. And § 101(a) (2) provides: Every member of any labor organization shall have the right to meet and assemble freely with other members; and to express any views, arguments, or opinions; and to express at meetings of the labor organizations his views, upon candidates in an election of the labor organization or upon any business properly before the meeting, subject to the organization's established and reasonable rules pertaining to the conduct of meetings: Provided, That nothing herein shall be construed to impair the right of a labor organization to adopt and enforce reasonable rules as to the responsibility of every member toward the organization as an institution and to his refrain- ing from conduct that would interfere with its performance of its legal or con- tractual obligations. 103. The Wagner Act, 49 Stat. 449 (1935), placed no restrictions on the negotiation of union security agreements. Unions and employers could make a closed shop contract. And where such a contract was made, an applicant for work would have to be a member of the union to qualify for employment. Furthermore, under a closed shop or a "common law" union shop an employee had to remain in good standing in his union to keep his job. Expulsion from the union for any reason, such as disagreement with the leadership, meant loss of employment. The hardship of this to the dissenting employee lead to changes embodied in the Taft- Hartley Act, 61 Stat. 140, 141 (1947), as amended, 29 U.S.C. §§ 158(a) (3), 158(b) (5) (1958). This statute goes far to separate the job from union membership. Under a Taft- Hartley (or Railway Labor Act) union shop all an employee need do to protect his job is "tender the periodic dues and initiation fees uniformly required as a condition of acquiring or retaining [union] membership. . ." 61 Stat. 141 (1947), as amended, 29 U.S.C. § 158 (a) (3) (1958). The employee need not maintain good standing in the union. And of course, he need not be a member of the union at the time of his employment. The congressional hearings which lead to these Taft-Hartley changes focused on the job control problems discussed above. While there was some discussion of the use of funds obtained under a union shop for political purposes, it was neither extensive nor systematic. See, e.g., Hearings on S. 55 Before the Senate Committee on Labor and Public Welfare, 80th Cong., 1st Sess. 796-808, 1004, 1425, 1687, 2145 (1947) ; Hearings o'n H.R. 8 Before the House Committee on Education and Labor, 80th Cong., 1st Sess. 305 (1947). The hearings did not bring sharply into focus the issues which are posed by Hanson and Street. These issues received next to no attention when Congress in 1951 reversed itself and permitted the union shop on the railroads. See, e.g., S. RaP. No. 2262, 8 1st Cong., 2d Sess. (1950) ; H.R. REP. No. 2811, 81st Cong., 2d Sess. (1950). But see 96 CONG. Rac. 17049- 50 (1951). These issues were not seriously considered in 1958, when the- Senate 'turned down Senator Potter's amendment-one of many amendments-to'the Labor-Maiagement Re- porting and Disclosure Act of 1958, S. 3974, 85th Cong., 2d Sess. (1958) (Kennedy-Ives). Senator Potter's amendment, which was unrelated to the problems in the bill it sought to amend, provided, inter alia: Any member of a labor organization whose employment is conditioned upon such membership may file a petition with the Secretary [of Labor] requesting that 19611 GOVERNMENTAL ACTION for the consequences of its new legislation upon the balance of power between 1 0 4 labor and management.

IV Even if all this is granted, however, it may very well be insisted that by adopting a slightly different form of words Congress might have made it un- avoidable for the Court to engage in constitutional review in cases like Street. For example, Congress might have passed a statute making union membership a condition precedent to employment on the railroads. The Court would then have had the inescapable obligation in a case posing the first amendment question of deciding the questions posed, no matter how difficult the task, and how much better Congress might be suited to solving the problem. It is true, too, that from the point of view of a dissenting employee, who is compelled to join a union, it is a matter of indifference whether the compulsion arises from a "command-type" or "permissive" statute. The short answer to this suggestion is that in view of the representation that minority employee interests enjoy in Congress it is most unlikely that such a statute would ever pass both houses. The longer answer requires recognition of the practical differences in the congressional approaches to "commanding" and "permissive" legislation, and of the nature of the traditional relationship between Congress and the Court.

moneys paid as membership dues or fees to such labor organization be expended exclusively for collective bargaining purposes or purposes related thereto. 104 CONG. REc. 11330 (1958) (vote recorded at 11347). The legislative history of the Labor-Management Reporting and Disclosure Act of 1959 shows that there was some concern with the Hanson-Street problem. See, e.g., 1 NLRB LEGISLATIVE HISTORY OF THE LABOR-MANAGEMENT REPORTING AND DISCLOSURE AcT OF 1959, at 481-82 (1959). Indeed, the Street case itself was called to the attention of the Senate. 2 id. at 1291-93. But this was strictly a side show and in no way central to the main legislative production which resulted in the 1959 statute. The Corrupt Practices Act, LMRA § 304, see note 83 supra, shows that Congress can be counted on to take action on the kinds of problems raised by the Hanson and Street cases. In United States v. CIO, 335 U.S. 106, 115 (1948), the Court isolated the stimuli to which Congress had responded when it extended corrupt practices legislation to labor unions: It was felt that the influence which labor unions exercised over elections through monetary expenditures should be minimized ... and that it was unfair to individual union members to permit the union leadership to make contributions from general union funds to. a political party which the individual member might oppose. The history of congressional regulation of union security and corrupt practices make it reasonable to assume that Congress will do what it has not yet done-systematically consider the problems which emerge from Hanson and Street, and take action necessary to remedy the situation. 104. Even the Labor-Management Reporting and Disclosure Act of 1959, which in Title VII amended the National Labor Relations Act, as amended, 61 Stat. 136 (1947), 29 U.S.C. §§ 151-68 (1958), gives a little to the union while imposing new restrictions on union organizational activity. Compare LMRDA § 702, with LMRDA § 704(c). THE YALE LAW JOURNAL [Vol. 70:345

In brief, congressional passage of a "command-type" statute would entail a fact-finding and deliberative process which would focus congressional attention on the kind of problem that occasioned Street-alleged use of union funds to compel ideological conformity within the membership. Two consequences would attend congressional passage of a "command-type" statute: first, in all likelihood Congress would recognize such problems, and would then accom- modate individual rights within the statutory framework; second, if a case requiring judicial review were to arise, the Court would have the benefit of a full congressional history-a history which would assist the Court in performing its judicial task. A specific example will aid an understanding of the differences between "command-type" and "permissive" legislation. If Congress itself were directly to establish wages, hours, and terms and conditions of employment for railroad workers the deliberative process in the legislature would be very different from what it was when Congress decided that collective bargaining between unions and employers over these matters should be encouraged by federal legislation. 0 5 For legislators are far likelier to focus hard on the impact their actions will have on individuals when new "commanding" legislation is in issue than when they contemplate simply permitting or encouraging private organizations to go their traditional ways and to assume initial, principal responsibility for decisions affecting the interests of those whom they represent. 106 When Congress repealed federal prohibition of the railroad union shop, and provided that union shop agreements were a proper subject for collective bargaining, the Congress was involved in the sort of process in which it had engaged when it decided that negotiations between unions and employers over wages, hours, and terms and conditions of employment were to be encouraged. Congress was not forcing anything upon anyone; it did no more there than allow the institution of collective bargaining to go its own free and traditional way.1° 7 The legislative process involved in this decision was one not calculated to develop or clarify a full panoply of problems similar to those involved in Street. This is a crucial difference between legislation which commands and legislation which permits.

105. Railway Labor Act § 2 Fourth, as amended, 48 Stat. 1187 (1934), 45 U.S.C. § 152 Fourth (1958). 106. "Collective bargaining [the labor policy which Congress did in fact legislate] was not defined by the statute which provided for it, but it generally has been considered to absorb and give statutory approval to the philosophy of bargaining as worked out in the labor movement in the United States." Order of R.R. Telegraphers v. Railway Ex- press Agency, Inc., 321 U.S. 342, 346 (1944). 107. The present prohibitions against all forms of union security agreements and the check-off were made part of the Railway Labor Act in 1934. They were enacted into law against the background of employer use of these agreements as devices for establishing and maintaining company unions, thus effectively depriving a substantial number of employees of their right to bargain collectively.... Since the enactment of the 1934 Amendments company unions have practically disappeared. S. RaP. No. 2262, 81st Cong., 2d Sess. 2-3 (1950). 19611 GO VERNMENTAL ACTION

Should not the factors which distinguish "permissive" from "command" legislation influence the Court to resist applying the Constitution to private action taken in response to a permissive statute? Where it can reasonably be anticipated that Congress will be sensitive to claims of oppression by dissenting members of private organizations, should not the Court be very reluctant indeed to regulate with so crude an instrument as the Constitution ?10s Is this not particularly true where the case involves political and economic complex- ities, and where the Court's isolation and institutional disabilities make an informed judgment unusually difficult ?109 And under these circumstances is not the course of restraint plainly correct when extension of judicial doctrine necessary to justify use of the Constitution will have implications in areas unrelated to the particular substantive problems before the Court? This latter consideration, of course, is present whenever the doctrine of governmental action is extended.110 Juxtaposed against these considerations is the plight of the injured plaintiff and others similarly situated. But this factor must be judged in conjunction with the wider ramifications of any one case-the effect of extending existing notions of governmental action. The plaintiff, it must be remembered, is asking that existing doctrine be extended."' Though the Court must be sensitive to the individual's hardship,11 2 it must consider as well, before passing on plaintiff's prayer, factors which transcend plaintiff's difficulties.

V These factors should have led the Court in the Hanson case to hold only that: 1) The union shop provision of the Railway Labor Act is a valid exercise of congressional power under the commerce clause. 2) It violates neither the first nor fifth amendments for Congress not to prevent employers and unions from negotiating union shop agreements. In these terms, Street raises the same problems as Hanson and should be decided the same way. The Street case affords the Court the happy opportunity to re-examine its earlier, and somewhat less than thoughtful, consideration of the doctrine of governmental action as it applies to labor unions.

108. Even when Congress does command there are doctrines of statutory construction by which the Court gives the Congress a chance to reconsider legislation of doubtful constitutional validity. Construing a statute to avoid a serious constitutional question-often in the teeth of other canons of construction-is a familiar rule designed to accomplish this purpose. The rule lets Congress re-examine a policy which in application has unfortunate effects on individuals--effects which were perhaps unforeseen by the legislators. Cf. Bickel & Wellington, Legislative Purpose and the Judicial Process: The Lincoln Mills Case, 71 HARV. L. REv. 1, 31-34 (1957). 109. See text at notes 76-78 supra. 110. See text at note 63 supra. 111. See te-xt at notes 63-64 supra. 112. See text at notes 63-64 supra. THE YALE LAW JOURNAL [Vol. 70: 345

VI Many of the factors which in combination press for this conclusion in Han- son and Street, are absent when the issue is whether the Constitution should be used to compel Lee Oliphant's admission to the Brotherhood of Locomotive Firemen and Enginemen. Realistically, there is little likelihood that Congress in the foreseeable future will pass a statute explicitly requiring collective bargaining agents to take to membership all the employees they represent, regardless of race. Negro employees are one minority inadequately represented in Congress. One of the least remarked aspects of the Labor-Management Report- ing and Disclosure Act of 1959, which contains a members Bill of Rights and comprehensively regulates union discipline, elections, and trusteeships, is the outrageous omission of an express provision requiring unions to admit all 11 3 employees to membership regardless of race. Moreover, there is little need in this situation to worry about the institutional capacity of the Court to solve Oliphant's problem through constitutional ad- judication. First, the reasonableness test is adequate to the purpose.114 Second, a judgment requiring admission of Negroes would have next to no implications for labor-management relations.11 5 It would neither increase nor decrease the power of unions vis-5.-vis employers. Of course, one problem does still remain: extension of the governmeital action doctrine to help Oliphant would make it conceptually difficult to contain the doctrine. With the major doctrinal jump already made, it would be im- possible to justify failure to extend it to the Hanson-Street problem, and perhaps to other, unrelated areas, where the uncertain consequences could be unfortunate." 6 Fortunately, Oliphant's problems can be solved in another way. VII Steele v. Louisville & N.R.R.,1" which established a union duty of fair representation, was decided in 1944. It held that the Railway Labor Act prevented unions from discriminating in collective bargaining against nonmember employees solely because they were Negroes. The Railway Labor Act says nothing about this duty. The Court, however, found it by implication and necessity from the right of exclusive representation granted to unions by the statute." 8

113. See 2 NLRB, LEGISLATIVE HISTORY OF THE LABOR-MANAGEMENT REPORTING AND DiscLOsuRE AcT OF 1959 1648-51 (1959) ; Givens, The Enfranchisement of Employees Arbitrarily Rejected for Union Membership, 11 LAB. LJ. 809 (1960). 114. See note 5 supra. 115. The union as has been noted again and again throughout this paper represents all employees whether or not they are members. See J. I. Case Co. v. NLRB, 321 U.S. 332 (1944); Order 'of R.R. Telegraphers v. Railway Express Agency, Inc., 321 U.S. 342 (1944). 116. See text at note 63 supra. 117. 323 U.S. 192 (1944). 118. Unless the labor union representing a craft owes some duty to represent non- union members of the craft, at least to the extent of not discriminating against them 1961] GOVERNMENTAL ACTION

But unions which refuse to admit Negroes continue to discriminate in collective bargaining.11 9 This has been established by a continuous line of cases since 1944.120 One reason for continued discrimination is not hard to find: an employee who is not a union member has no voice in the political processes of the union. This means that he cannot participate in the choice of leadership. It also means that he cannot contribute to the formulation of the union's collective bargaining policies, nor vote to approve or reject contracts which will 121 regulate his industrial life.

as such in the contracts which it makes as their representative, the minority would be left with no means of protecting their interests, or indeed, their right to earn a livelihood by pursuing the occupation in which they are employed. While the majority of the craft chooses the bargaining representative, when chosen it represents, as the Act by its terms makes plain, the craft or class, and not the majority. The fair interpretation of the statutory language is that the organization chosen to represent a craft is to represent all its members, the majority as well as the minority, and it is to act for and not against those whom it represents. It is a principle of general application that the exercise of a granted power to act in behalf of others involves the assumption toward them of a duty to exercise the power in their interest and behalf, and that such a grant of power will not be deemed to dispense with all duty toward those for whom it is exercised unless so expressed. Steele v. Louisville & N.R.R., 323 U.S. 192, 201-02 (1944). The duty is found in the Railway Labor Act § 2 Fourth, as amended, 48 Stat. 1187 (1934), 45 U.S.C. § 152 Fourth (1958), and in the National Labor Relations Act § 9(a), as amended, 61 Stat. 143 (1947), 29 U.S.C. § 159(a) (1958). 119. The court that decided Steele did not know this would be so. And thus in 1944 Stone, C.J., for the Court was able to say that the union had "the right to determine eligibility to its membership ... ." 323 U.S. at 204. But even in 1944 it was also said that "the union is required to consider requests of non-union members of the craft and expres- sions of their views with respect to collective bargaining with the employer and to give to them notice of and opportunity for hearing upon its proposed action." Ibid. These are words without substance in the absence of membership, for union leadership bent upon discrimination is not likely to give any weight to the minority employee who has no political power in the organization because he has no vote. 120. Every indication is that so long as Negroes are excluded from membership the unions will not represent them fairly, as required by the statute.... The Court's decision in . . . Steele . . . [has] not put an end to racial discrimination by the BLFE [Brotherhood of Locomotive Firemen and Enginemen]. For further instances see Hinton v. Seaboard Air Line R.R., 23 L.R.R.M. 2097 (4th Cir. 1948) ; Graham v. Southern Ry., 74 F. Supp. 663 (D.D.C. 1947) ... '. Aaron & Komaroff, Statutory Regulation of Internal Union Affairs-I, 44 ILL. L. REv. 425, 436 & n.56 (1949). " There have been many cases indeed since Steele where a Iiure-white union has dis- criminated in bargaining against Negro employees. See, e.g., Graham v. Brotherhood of Locomotive Firemen, 338 U.S. 232 (1949) ; Brotherhood of R.R. Trainmen v. Howard, 343 U.S. 768 (1952); Clark v. Norfolk & W. Ry., 37 L.R.R.M. 2685 (W.D. Va. 1956). 121. Nothing in the Labor-Management Reporting and Disclosure Act of 1959 has changed the law in this regard. THE YALE LAW JOURNAL [Vol. 70:345

Membership is by no means a guarantee of fair representation. Majorities discriminate against minorities. 12 2 But experience and available evidence sug- 12 3 gest that there can never be fair representation without membership. Under these circumstances the statutory duty of fair representation should be interpreted to require unions to admit Negro employees to membership. The alternative is for pure-white unions to stop serving as statutory exclusive bargaining representatives. 124 If Lee Oliphant has a cause of action under the 125 Railway Labor Act, he does not need the fifth amendment.

CONCLUSION The easy conclusion, shared by too many "bold thinkers," that "whenever any organization or group performs a function of a sufficiently important public nature, it can be said to be performing a governmental function and thus should have its actions considered against the broad provisions of the Consti- tution" is wrong.'2 Like most easy conclusions about most hard governmental problems it lacks the institutional feel. Perhaps there are private groups in society to which the Constitution should be applied.' 2 7 But one thing is clear: that conclusion should depend on more than an awareness that the group commands great power or performs a function of an important public nature. With respect to labor unions one should ask: (1) What is the present state of the law? (2) Where will an extension of doctrine lead? (3) Can noncon- stitutional law be used to solve the problem? and (4) Can and will Congress solve these problems-and solve them better than the Supreme Court? The answer to these four questions yields one conclusion: union conduct should not now be regulated by the Constitution of the United States. 28

122. See, e.g., Jennings v. Jennings, 91 N.E.2d 899 (Ohio Ct. App. 1949) ; Conley v. Gibson, 355 U.S. 41 (1957). 123. See note 119 supra. 124. Cf. Betts v. Easely, 161 Kan. 459, 169 P.2d 831 (1946). 125. The Court could not solve the Hanson-Street problem by such statutory interpretation. Even if the statute could be twisted to permit the Court to regulate the union's use of dues for political purposes, the Court should not undertake the task. The reasons which militate against using the Constitution to regulate the political activities of union, apply in large part to Court development of a statutory program in the absence of legislative guidance. Unlike Oliphant's problem, which the Court is institutionally well quali- fied to solve, see text at notes 68-72 supra, the supervision of union expenditures which the Hanson-Street situation involves wvould plunge the Court into problems which it is institutionally ill-equipped to handle. Congress is capable of developing a program to super- vise union expenditures, and can be expected to do the job. 126. Miller, Priiate Governments and the Constitution 11 (Occasional Paper, Center for Study of Democratic Institutions, 1959). 127. To my mind the Jaybird Democratic Association of Texas is inescapably such a group. See Terry v. Adams, 345 U.S. 461 (1953). 128. Compare Cox, The Role of Law in Preserveing Union Democracy, 72 HARV. L. REv. 609, 620 (1959). It has also been argued that the powers which the NLRA vests in labor unions are so far governmental that all actions, including the election and rejection of mem- 1961] GOVERNMENTAL ACTION 375

bers, are subject to the restrictions which the fifth and fourteenth amendments impose upon federal and state authorities.... In my opinion the reasoning is highly dangerous. The implications of calling labor unions governmental instrumentalities are not easy to perceive, but surely the designation would invite more and more regulation with consequent loss of independence. This paper has not dealt with the whole range of additional considerations suggested by the last portion of the quotation from Professor Cox. But see Wellington, Union Democracy and Fair Representation: Federal Responsibility in a Federal System, 67 YALE L.J. 1327 (1958), for a discussion, in a different context, of some of these considerations.